Alcohol problems among American Indians have received the lion 's share of research attention for good reason. Alcohol abuse and dependence are, without question, among the most significant public health problems in American Indian communities and understanding the causes and consequences of these disorders has been an important goal. Yet, in the rush to comprehend the difficulties that Indian people have with alcohol, researchers have also run the risk of losing sight of the strengths that Indian communities show when dealing with these same problems. This project seeks to explore some of these positive aspects of the American Indian experience with alcohol through a thorough exploration of the little understood, but often quite high, rates of remission from alcohol problems among Indian people in later life. It is hoped that by better understanding these dynamics in a random community sample of Indian people in the northern plains, this project will yield information that can be immediately applied in programmatic community efforts to address alcohol problems. Toward those ends, this project is proposed in partnership with the tribal council and its alcohol program and the research methods incorporate community input throughout to inform the design and conduct of the research as well as the interpretation of its findings. The project has four specific aims: 1) to investigate how problem drinking and abstinence are meaningfully construed and experienced by American Indian men and women; 2) to determine the factors that are relevant to understanding how alcohol-dependent men and women give up drinking and maintain their abstinence; 3) to establish a typology of pathways to abstinence in this community; and 4) to link characteristics of American Indian individuals with their pathways to abstinence. This research will employ a sample of all community members who met the lifetime criteria for alcohol dependence (DSM III-R/DSM-IV) in a prior study of psychiatric epidemiology in the same community. Preliminary data from this random community sample suggest that one half of those with a lifetime history of alcohol dependence did not meet the criteria for alcohol dependence in the year prior to the original psychiatric epidemiological interview. The goal of this research is to employ both qualitative and quantitative methods to understand this dynamic in order to ensure the validity of what is measured, by grounding measurements in local experience, and to rigorously test, using a careful measurement strategy, the hypotheses that emerge in the course of the qualitative inquiry.